ClinSQL - Laboratory Results Analysis: leaderboard
Metric: Execution Score (%). Source: arxiv.org. Saturation forecast: Around December 2026. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek R1 0528 | 66.91 |
| 2 | GPT-5 Mini | 65.09 |
| 3 | Gemini 2.5 Pro | 63.66 |
| 4 | GPT-5 Chat | 57.89 |
| 5 | Qwen 3 235B A22B 2507 Instruct | 55.94 |
| 6 | Gemini 2.5 Flash | 54.4 |
| 7 | DeepSeek V3.1 | 53.66 |
| 8 | Qwen 3 Coder 480B A35B Instruct | 49.78 |
| 9 | GPT-5 Nano | 49.02 |
| 10 | O4 Mini (2025-04-16) | 48.45 |
| 11 | GPT-4.1 | 48.39 |
| 12 | Grok 4 Fast (Reasoning) | 45.5 |
| 13 | Qwen 3 Next 80B A3B Instruct | 42.33 |
| 14 | Llama 4 Maverick Instruct | 37.66 |
| 15 | Mistral Medium 3 | 32.22 |
Interactive version: theaggregate.ai/benchmark?slug=clinsql-laboratory-results-analysis · How It Works · Data refreshed daily, snapshot 2026-09-25.